<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
all-base-guten-rarity-all-2p5k-rerun
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.1039
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.331 | 0.29 | 500 | 5.3409 |
| 5.0388 | 0.59 | 1000 | 4.9224 |
| 4.6975 | 0.88 | 1500 | 4.6894 |
| 4.4365 | 1.17 | 2000 | 4.5460 |
| 4.2942 | 1.47 | 2500 | 4.4316 |
| 4.1878 | 1.76 | 3000 | 4.3204 |
| 4.06 | 2.05 | 3500 | 4.2464 |
| 3.884 | 2.35 | 4000 | 4.1993 |
| 3.8643 | 2.64 | 4500 | 4.1495 |
| 3.8236 | 2.93 | 5000 | 4.0953 |
| 3.6231 | 3.23 | 5500 | 4.0915 |
| 3.5763 | 3.52 | 6000 | 4.0668 |
| 3.5604 | 3.81 | 6500 | 4.0295 |
| 3.4609 | 4.11 | 7000 | 4.0327 |
| 3.3132 | 4.4 | 7500 | 4.0270 |
| 3.3027 | 4.69 | 8000 | 4.0148 |
| 3.2932 | 4.99 | 8500 | 4.0008 |
| 3.1357 | 5.28 | 9000 | 4.0145 |
| 3.1262 | 5.57 | 9500 | 4.0132 |
| 3.1241 | 5.87 | 10000 | 4.0119 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
